Search for Vector-like Top Quarks in di-Leptons

Anthony Barker

July 29, 2017 DPF 2017

1

What is a vector-like quark• They are not actually vector particles; they’re spin 1/2 fermions.

• They’re “vector-like” in that left and right helicity components have the same quantum numbers and transform identically under gauge transforms.

• No Yukawa coupling (y) They don’t get their mass from the Higgs

• They are typically an expansion of the third generation rather than a 4th generation (T mixes with top), so they’re not excluded by 4th gen searches.

v is the Higgs VEV

Anthony Barker

m ̄

2

• Largely model independent analysis• Few free parameters. • VLQ’s are the simplest type of colored fermions still experimentally

allowed.

Motivation

Aguilar–Saavedraa, Benbrikc, Heinemeyer, Pérez-Victoria arXiv 1306.0572

Theory• VLQ’s are REQUIRED if the Higgs is a pseudo-Goldstone boson, and

in the partially-composite theories of flavor, including little Higgs and composite Higgs models. ➡In both mechanisms, VLQ have sizable mixings with 3rd

generation SM quarks, hence top and bottom partners.

Observation

Analysis

• VLQs can improve the fit of electroweak observables to data:

�2SM = 10.97,�2

V LQ = 1.61

Anthony Barker3

T 2/3

B�1/3

Y �4/3

X5/3

DecaysParticleElec. Charge

The cast of characters:

TB

T, BT, XB, Y

T, B, XT, B, Y

Singlets

Doublets

Triplets

7 Allowed Combinations

Anthony Barker

tW+

bW�

bW+, tH, tZ

tW�, bH, bZ

4

What they look like• VLQ can be produced singly or in

pairs. • Pair production is insensitive to

couplings. • Limits depend on the branching

ratios—a function of multiplet and mass.

• Complicated decays! Need multiple approaches and combined results.

1311.7667 and 1311.7667

50%

25%

Anthony Barker 5

T’ decay and Feynman diagrams

BR

(T→

tH)

BR(T→bW)BR(T

→tZ

)

Singlet

Doublet

T Mass Limit Plane

Complex decays, thousands of final states

Need multiple analyses combine

3 corners, many possible approaches

Anthony Barker

6

Last Year’s Limits

Excluded by tH search

Extant

M(T) ≤ 850 GeV Fully Excluded

0 0.2 0.4 0.6 0.800.20.40.60.8

Unphysical

= 1050 GeVTm

0 0.2 0.4 0.6 0.8

Unphysical

= 1100 GeVTm

0 0.2 0.4 0.6 0.8 1

Unphysical

= 1150 GeVTm

0.20.40.60.8

Unphysical

= 900 GeVTm

Unphysical

= 950 GeVTm

Unphysical

= 1000 GeVTm

0.20.40.60.8

Unphysical

= 750 GeVTm

Unphysical

= 800 GeVTm

Unphysical

= 850 GeVTm

0.20.40.60.8

Unphysical

= 600 GeVTm

Unphysical

= 700 GeVTm1

Wb)→BR(T

Ht)

→BR

(T

ATLAS Preliminary-1 = 13 TeV, 13.2 fbs

SU(2) doublet SU(2) singlet

Exp. limit Obs. limitHt+X 1-lepHt+X 0-lepHt+X CombinationSinglet Excluded

below 1000 GeVDoublet Excludedbelow ~1150 GeV

Incomplete list

Anthony Barker7

with ad-hock CMS & ATLAS overlays

Excluded by bWbWsemi-leptonic search

Pure bW Excluded below 1300 GeV

Excluded by 1 lep + MET searchExcluded by bW+hadrons search

Dilepton Channel

Part of a program of several lepton analyses

5% branching fraction of TT goes to dileptons

Anthony Barker8

Di-Lepton Selection

! OSDL1 ChannelTargets bWbW! 2 Tight OSDL! Z veto (76-106 GeV)! 2-3 Jets! ≥ 1 B-jet! ST > 1000 GeV! Min Mlb > 170 GeV

Top Rejection! HT > 325 GeV

! MET > 30 GeV,

! Mll > 20 GeV

! SSDL ChannelTarget tH+X! 2 Tight SSDL! ≥ 3 Jets! ≥ 1 B-jet! ST > 1100 GeV! HT > 500 GeV

! MET > 30 GeV

! Mll > 20 GeV

Ele Pt threshold: 30 GeVMuon Pt threshold: 30 GeV

Anthony Barker

9

charge misID rate.

Anthony Barker10

Fake Rate

Anthony Barker

Prompt Rate

Fake rate measured in Lep+Jet data

Prompt and charge flip measured in DY data

Charge-flip Rate

Measured inputs to data driven background method

11

Work in Progress

Result Plots• Significant reliance on MC

• 12 systematic uncertainties just from MC

• 17 systematic uncertainties total

• Prescaled 8x for partial blinding

Anthony Barker12

Back-up

13

CMS &ATLAS8 TeV Limits

White = Extant

CMS Exclusion Overlay

ATLAS Results page

CMS B1/3: B2G-13-006CMS T2/3: B2G-13-005

T2/3

B-1/3

Anthony Barker14

Branching fractions of TT

• Analyzed with Particle Branch Ratio Counter

• Full decay tree explored, including all permutations of final states.

• tH(WW) is necessary on at least one branch for SSDL

• second lepton usually from a top, or W on the other branch.

Branching fractions in TT

Proportion of each column by source

15

Electrons• Pt > 30, |Eta| < 2.4• mini-iso < 0.1 for tight• mini-iso < 0.4, used for

background studies• Custom MVA working point• Electron scale factors

Anthony Barker

Muons• Pt > 30, |Eta| < 2.4• mini-iso < 0.1 for Tight• mini-iso < 0.4 for background

studies• Muon scale factors

16

Systematic Uncertainties• All of the following systematic uncertainties are

included and fully implemented as shape uncertainties: • Renormalization scale uncertainty• Parton Shower E Scale uncertainty• PDF uncertainty• Jet scale factor uncertainty• Jet Energy Scale uncertainty• Jet energy resolution uncertainty• B-tagging scale factor uncertainty • light-jet ID scale factor uncertainty• HLT• Lepton ID SF uncertainty• Lepton Iso SF uncertainty• Top Pt reweighting

• Simple constant uncertainties: • Lumi uncertainty

• Bin-by-bin• Statistical (on plots)• Prompt rate

uncertainty• Fake rate uncertainty • Charge flip

uncertainty

17